AQUATIC BIOLOGY Aquat Biol
Vol. 1: 225–238, 2008 doi: 10.3354/ab00024
Published online January 15
OPEN ACCESS
Bacterial symbionts and mineral deposits in the branchial chamber of the hydrothermal vent shrimp Rimicaris exoculata: relationship to moult cycle Laure Corbari1,*, Magali Zbinden2, Marie-Anne Cambon-Bonavita3, Françoise Gaill2, Philippe Compère1 1
Université de Liège, Laboratoire de Morphologie Fonctionnelle et Évolutive, Unité de Morphologie Ultrastructurale, Allée de la Chimie, 3, 4000 Liège, Belgium
2
UMR CNRS 7138 ‘Systématique, Adaptation et Evolution’, Université Pierre et Marie Curie, 7 Quai St Bernard, Bâtiment A, 75252 Paris Cedex 05, France 3 Laboratoire de Microbiologie et Biotechnologie des Extrêmophiles, IFREMER, Centre de Brest, BP 70, 29280 Plouzané, France
ABSTRACT: The shrimp Rimicaris exoculata is considered a primary consumer that dominates the fauna of most Mid-Atlantic Ridge hydrothermal ecosystems. The shrimps harbour in their gill chamber an important ectosymbiotic community of chemoautotrophic bacteria associated with iron oxide deposits. The settlement and development of this ectosymbiosis was investigated using microscopy techniques (light microscopy, LM; and scanning, transmission and environmental scanning electron microscopy: SEM and ESEM, respectively) for shrimps from 2 different vent fields (Rainbow, 36° 14.0’ N and TAG, 26° 08.0’ N). The results revealed a bacterial re-colonisation after each exuviation and a development of the bacterial community in 5 steps in relation to the moult stages, which were used as a reference time scale. In 287 shrimps from both vent fields, pre-ecdysial stages prevailed in the population, suggesting a short anecdysis and high moulting rate, probably to renew the ectosymbiosis. Comparisons with moult cycles of littoral shrimps suggest that the interval between successive exuviations in R. exoculata may be as short as 10 d. The colours of R. exoculata result from accumulation of iron oxide, which forms a bacteria-associated mineral crust in the gill chambers. The close correspondence between moult stages, the development of the ectosymbiont community and shrimp colours indicate that colour could be used to rapidly determine shrimp moult stages. KEY WORDS: Hydrothermal vents · Shrimp · Moult cycle · Ectosymbiosis · Iron oxides Resale or republication not permitted without written consent of the publisher
The discovery of deep-sea hydrothermal vents in 1977 revealed many new animals that derive their nutrition from chemoautotrophic bacteria, housed in a variety of tissues and structures (Cavanaugh et al. 2006). The caridean hydrothermal vent shrimp Rimicaris exoculata (Williams & Rona 1986) is one species that dominates the fauna at several Mid-Atlantic Ridge (MAR) vent sites. It forms large swarms on the chimney walls, reaching densities of up to 2500 ind. m–2 (Desbruyères et al. 2001). R. exoculata possesses an enlarged gill chamber, housing an abundant community
of bacteria, mostly distributed on bacteriophore setae of the shrimp mouthparts (i.e. scaphognathites, exopodites of the first maxilliped) and on the inner side of the gill chamber (Casanova et al. 1993). Although several morphotypes have been observed in electron microscopy, the bacteria are described as a monoculture of a single phylotype of epsilon-Proteobacteria (Polz & Cavanaugh 1995). Their appearance, abundance, and attachment to the cuticle suggest an ectosymbiosis (Van Dover et al. 1988, Casanova et al. 1993, Segonzac et al. 1993). Several studies (Wirsen et al. 1993, Polz & Cavanaugh 1995) suggesting that the bacterial symbionts acquire energy from sulphide oxidation
*Email:
[email protected]
© Inter-Research 2008 · www.int-res.com
INTRODUCTION
226
Aquat Biol 1: 225–238, 2008
have not been experimentally confirmed. The definitive role of these bacteria has still not been clarified, although ectosymbionts have often been considered part of the shrimp diet. Shrimp could feed directly on the bacterial ectosymbionts by grazing on them inside the gill chamber, as indicated by stable isotope measurements (Gebruk et al. 1993, Rieley et al. 1999), but the bacteria in the shrimp mid-gut have also been proposed as an alternative nutritional source (Pond et al. 1997, Polz et al. 1998, Zbinden & Cambon-Bonavita 2003). In previous studies the presence has been observed of red-brown mineral deposits in the gill chamber of Rimicaris exoculata, including mouthparts and branchiostegites (Gloter et al. 2004, Zbinden et al. 2004). These deposits have been identified as hydrous iron oxide in the form of ferrihydrite (Gloter et al. 2004). Because of their close association with the bacterial cell walls, these minerals have been hypothesised to result from bacterial metabolism, suggesting, therefore, the presence of chemoautotrophic iron-oxidisers among the bacterial community (Zbinden et al. 2004). Macroscopically, these iron oxides deposited in the gill chamber generate different shrimp colours (Zbinden et al. 2004). For instance, shrimps from the Rainbow vent site exhibit rusty or red hues, while those from the TAG vent site appear mostly dark or black and rusty brown (Gebruk et al. 1993, Zbinden et al. 2004). Changes in iron oxide organization, structure, and abundance are probably responsible for these colour changes. Zbinden et al. (2004) described 3 bacterial morphotypes (rods, thin and thick filaments) and mapped these bacteria and associated minerals within the whole gill chamber, including the mouthparts, by subdividing the chamber into 3 functional compartments
(Fig. 1), considered as distinct microenvironments: (1) the lower pre-branchial chamber, which houses bacteria but few minerals, (2) the true branchial chamber, which contains the gills but is free of bacteria and minerals, and (3) the upper prebranchial chamber, housing most of the bacteria and associated minerals. Zbinden et al. (2004) were the first authors to suggest that the abundance of bacteria and associated minerals could be related to the shrimp moult cycle, because the bacteria are directly attached to the cuticle which is shed at ecdysis and subsequently renewed, as is the case in all arthropods (Drach 1939; see review in Charmantier-Daures & Vernet 2004, Compère et al. 2004). Thus, bacterial re-colonisation must occur after each exuviation. The aim of the present study was to follow the bacterial re-colonisation of the gill chamber and the progress of the mineral–bacteria association between 2 successive moults, in Rimicaris exoculata shrimps from 2 MAR hydrothermal vent sites, Rainbow and TAG. These observations have been related to the moult stages and the external colour of the shrimps, in an attempt to provide a relative and/or an absolute time scale for bacterial colonisation and mineral deposition.
MATERIALS AND METHODS Shrimp collection. Specimens of Rimicaris exoculata (Williams & Rona 1986) were collected during the French ‘EXOMAR’ cruise (August 2005) to the Rainbow (36° 14.0’ N, 2300 m depth) and TAG (26° 08.0’ N, 3600 m depth) hydrothermal vent sites on the MAR. A suction sampler on the ROV (remotely operated
Fig. 1. Rimicaris exoculata. Colour patterns: (A) specimen with white branchiostegites; and (B) specimen with red branchiostegites. (C) Schematic view of the gill chamber showing the disposition of the scaphognathite (sc) and the exopodite of the 1st maxilliped (ex) as well as the location of its 3 functional compartments — (1) lower pre-branchial chamber; (2) true gill chamber facing the gills; (3) upper pre-branchial chamber. The dotted lines delimit the observed area (median zone) with predominant bacterial colonisation (redrawn from Zbinden et al. 2004)
Corbari et al.: Ectosymbiosis and moult cycle in vent shrimp
vehicle) ‘Victor 6000’ was used, operating from the RV ‘Atalante’. Immediately after retrieval, entire living specimens were dissected into body parts (e.g. branchiostegites, tail) and fixed onboard in a solution of 2.5% glutaraldehyde in a mixture of seawater (salinity 33 ‰) and freshwater (dilution 7/10 at pH 7.2). Moult stages and colour categories. The moult stages of 143 ind. of Rimicaris exoculata from Rainbow and 144 ind. from TAG were determined according to the Drach & Tchernigovtzeff (1967) moult-staging method, based on the development of setae matrices along the uropod borders and on the hardness of the new cuticle. According to the common nomenclature in use in decapod crustaceans (Charmantier-Daures & Vernet 2004), the moult cycle is subdivided into 3 periods: (1) the preecdysis, preceding the shedding of the old cuticle (Stages D0 – D1’–1’’–1’’’ and D2 –D4), (2) the postecdysis (Stages A/B, C) and (3) the anecdysis (Stage C4), which is a period of integument stability between 2 successive moults. Sampled shrimps exhibited different colours mainly due to red mineral deposits on the integument and especially in the gill chamber (Fig. 1A). The colour of the telson and uropods (forming the tail plate) was observed on all staged specimens and used to classify individuals into colour categories. Light and electron microscopic observations were performed on Rimicaris exoculata branchiostegites in an attempt to follow bacterial colonisation and mineral deposition through a moult cycle. In order to compare the bacterial cover in different shrimps, the median zone of the branchiostegite (Fig. 1C) was selected as reference area because it presents a regular bacterial cover and lines both the antero-ventral and anterodorsal compartments (Zbinden et al. 2004). Before preparation for electron microscopy, the complete branchiostegite and samples taken from it were observed and photographed under a stereo microscope (Olympus SZ40). Light microscopy (LM). Samples from the branchiostegites (median zone) were dissected from individuals representative of each moult stage and both vent sites (n = 15 from Rainbow and n = 20 from TAG). These samples were post-fixed in osmium 1%, dehydrated in an ethanol and propylene oxide series and then embedded in epoxy resin (SPI-PON 812). Semi-thin sections were obtained from a Reichert-Jung Ultramicrotome (Ultracut E) using a diamond knife, and were then stained with toluidine blue (pH 9.0) for observation using LM (Olympus SZ40). Scanning electron microscopy (SEM). Branchiostegite samples (median zone) were taken from shrimps from Rainbow (n = 10) and TAG (n = 15) selected as representative of the different moult stages. They were dehydrated through an ethanol
227
series, critical-point dried and platinum coated (20 nm) in a Balzers SCD-030 sputter-unit before observation using SEM (JEOL JSM-840A) at 20 kV accelerating voltage. Back-scattered electron imaging (BSE). A more precise investigation of the structure of bacteriaassociated minerals was carried out on polished thin slices (20 to 50 µm) of branchiostegites from Rainbow (n = 10) and TAG (n = 10) specimens. The samples were dehydrated in an ethanol and propylene oxide series, then embedded in epoxy resin (EpoFix, Struers) for geological specimens. Polished thin slices on glass slides were produced by abrasion on diamond disks and were finally mirror polished with a non-aqueous 1 µm diamond suspension (ESCIL, PS-1MIC). The polished thin slices were surrounded by conductive silver paint and then carbon-coated in a Balzers BAF-400 rotative evaporator. Structural observations of the bacteria-associated minerals were performed in an environmental scanning electron microscope (FEI XL30 ESEMFEG), working at 15 to 20 kV accelerating voltages. A total of 30 polished thin slices were imaged by secondary electrons (SE) and back-scattered electrons (BSE).
RESULTS Moult stage distributions To characterise the moult cycle of Rimicaris exoculata, the moult stages were determined on a total of 287 ind. (n = 143 from Rainbow site and n = 144 from TAG site). Sampling is supposed to be random and the massive shrimp collection can be considered as representative of the population. The percentages of individuals (% ind.) at each moult stage represent their relative proportions within the population and should correspond to the relative duration of the stages in the moult cycle (see ‘Discussion’). Results obtained from both Rainbow and TAG populations were pooled because no significant difference was observed between them. Fig. 2A,B shows that more than 80% of the shrimps were in preecdysis (Stages D0 –D4) with a maximum of 50% of the individuals at Stage D1, the longest moult stage. In contrast, anecdysis (Stage C4) accounted for
